Data-Driven Application Maintenance: Views from the Trenches

TL;DR

This paper shares practical insights from implementing a machine learning-based application maintenance system that tackles common IT service issues, emphasizing pragmatic, user-friendly solutions for real-world adoption.

Contribution

It introduces a data-driven, machine learning approach for application maintenance that addresses real-world challenges with low barriers to adoption and tailored complexity.

Findings

Successful implementation of a proof of concept for incident management tasks

Identification of key barriers to adopting research in practice

Emphasis on pragmatic solutions aligned with business constraints

Abstract

In this paper we present our experience during design, development, and pilot deployments of a data-driven machine learning based application maintenance solution. We implemented a proof of concept to address a spectrum of interrelated problems encountered in application maintenance projects including duplicate incident ticket identification, assignee recommendation, theme mining, and mapping of incidents to business processes. In the context of IT services, these problems are frequently encountered, yet there is a gap in bringing automation and optimization. Despite long-standing research around mining and analysis of software repositories, such research outputs are not adopted well in practice due to the constraints these solutions impose on the users. We discuss need for designing pragmatic solutions with low barriers to adoption and addressing right level of complexity of problems with respect to underlying business constraints and nature of data.